Aristi et al. (2015) analysed the effects of effluent from a wastewater treatment plant on river biofilms and ecosystem metabolism, by comparing a segment of the river above the WWTP effluent with three stations influenced by the pollutants. Concentrations of common pharmaceuticals (diclofenac, ibuprofen, carbamazepine, venlafaxine) ranging from a few to tens of ng/L for each compound were measured in all segments of the river. The analyses showed that effluent acted as a subsidy on biofilm biomass and ecosystem respiration (Figure 2), probably as a consequence of enhanced availability of organic carbon. This effect showed a high correlation with both changes in nutrient (nitrate and phosphate) and pharmaceutical concentrations. However, the study also observed some indications of stress effects, which were reflected in non-photochemical quenching disturbances in the biofilms. Chonova et al. (2018) also conducted an in-depth analysis of the effects of pharmaceuticals emitted from the WWTP on river biofilm by comparing communities of biofilm-forming bacteria above and below the effluent inflow from a WWTP. The 3-years study analysed environmental samples during summer and winter and determined concentrations of 10 commonly used pharmaceuticals. It was observed that bacterial community structures, characterized by denaturing gradient gel electrophoresis (DGGE), were strongly related to concentrations of sulfamethoxazole for the station located close downstream to the effluent (Table 1). On the other hand, DGGE profiles from the station located far downstream from the effluent were related to concentrations of ibuprofen and atenolol. Aubertheau et al. (2017) also examined the effect of WWTP effluent on river biofilm. Their study analysed 12 pharmaceuticals from different groups at sites located downstream from 12 WWTPs, showing the presence of 11 of them in the biofilm, with the highest occurrence frequency for diclofenac, carbamazepine, sulfamethoxazole and propranolol (Table 1). The results confirm that biofilm-forming microorganisms can sorb and accumulate numerous pharmaceuticals. In addition, this work shows that WWTP effluent can be a factor modifying biofilm species composition, in particular reducing the abundance of the cyanobacteria present (Figure 2). Importantly, the abundance of resistance genes was observed to increase. Antimicrobial Resistance (AMR) is a phenomenon that involves the activation or development of mechanisms enabling microorganisms to survive in the presence of chemicals intended to kill them (Christaki et al., 2020). The release of residual antimicrobial compounds from WWTPs into the environment is one important cause of this phenomenon. Observations in recent years unambiguously confirm that AMR should be counted among the main threats to the modern world, and this phenomenon, although already causing a significant problem, is likely to intensify (O’Neil, 2014; Matviichuk et al., 2022).

Rosi-Marshall et al. (2013) used PhaDS filled with selected compounds at concentrations of 12–15 mM each, to investigate the effects of several pharmaceuticals (caffeine, ciprofloxacin, cimetidine, diphenhydramine, metformin, and ranitidine) individually and in mixtures on river biofilm. Of the three antihistamine compounds tested, only diphenhydramine had a statistically significant effect on gross primary production and respiration, causing rates of both processes to decrease significantly (Table 1). Interestingly, the effects of this compound were only slightly smaller than those of the antibiotic, ciprofloxacin. Effects on respiration were observed on both the inorganic substrate at the top of the PhaDS device, which promotes the growth of photosynthetic organisms, and the organic substrate, which promotes the growth of heterotrophic bacteria. Analysis of biofilm composition indicated that the observed diphenhydramine effect on respiration may be due to changes in species composition. Similar observations were noted in the work of Rosi et al. (2018), where they compared the effects on biofilm in streams with different degrees of urbanization, observing that the effects of ciprofloxacin and diphenhydramine on respiration occurred in the least urbanized sites, while they were not observed in the more polluted ones, suggesting the development of tolerance to these pharmaceuticals when there is constant exposure. Interestingly, in the case of ciprofloxacin, it was observed that in the urban streams where this compound had no significant effect on respiration, it caused changes in the taxonomic composition of the biofilm. Furthermore, the effect on composition was greatest at the site where respiration was least affected. Shaw et al. (2015) studied the same pharmaceuticals except that the diffusing substrates were placed in lentic systems. In this work, a strong effect of diphenhydramine on biofilm-forming organisms was observed, where the decrease in rates of gross primary production and respiration was much greater than for other tested compounds, including ciprofloxacin. Ogata et al. (2020) observed that a mixture of diphenhydramine and caffeine affected the taxonomic composition of stream biofilms (Table 1 ), although this was primarily through promoting unique taxa associated with contaminant tolerance and/or degradation, as caffeine and diphenhydramine, may have served as a carbon and/or energy source for taxa capable of degrading one or both compounds.

Several studies have evaluated the effects of pharmaceutical mixtures on biofilm-forming microorganisms under controlled laboratory conditions. Proia et al. (2013) conducted translocation experiments in which they exposed the biofilm for 16 days to water taken from a clean reference river, a moderately polluted river and a heavily polluted river in which pharmaceuticals (57 compounds) and pesticides (16 compounds) were detected. After biofilm translocation from clean to polluted water, it was observed that autotrophic biomass and peptidase increased while phosphatase and photosynthetic efficiency decreased, and these effects were associated with analgesics and anti-inflammatories. In contrast, Corcoll et al. (2015) used artificial streams to evaluate the effects of a mixture of nine pharmaceuticals at 5 μg/L on river biofilm using fully controlled conditions. During 11 days of experiment, pharmaceuticals caused a decrease in biomass and number of algal species, as well as changes in bacterial structure, as demonstrated by a reduction of the operational taxonomic unit (OTU) richness (Table 1). On the other hand, a stimulatory effect of the tested pharmaceuticals was also observed, which caused an increase in primary production and respiration, which could be related to the promoted growth of green algae. Using artificial stream mesocosms, Robson et al. (2020) analysed the effects of a mixture of three pharmaceuticals—fluoxetine, diphenhydramine and ciprofloxacin—on established and successional river biofilm for 20 days. Although the effect on established biofilm was small, a substantial reduction in both gross primary production and community respiration was observed in successional biofilm which was attributed to the lack of a protective EPS (extracellular polysaccharide) layer (Table 1). Under shaded conditions, these same compounds also interfered with denitrification. The major effect of these pharmaceuticals on the established biofilm was on changing in diatom community structure.

Several papers have highlighted the potential effects of SSRIs on freshwater microorganisms (Richmond et al., 2016; Richmond et al., 2019; Yang et al., 2019; Robson et al., 2020; Cui et al., 2021). Richmond et al. (2016) analysed the effects of fluoxetine and citalopram (20 μg/L of each; individually and in a mixture) using artificial stream mesocosms. Both compounds suppressed gross primary production and community respiration but had no effect on algal biomass and whole-stream metabolism (Table 1). Synergistic effects were not observed with the mixture, as results were similar to those in individual exposures. In a subsequent paper by Richmond et al. (2019), the same artificial stream mesocosms were exposed to fluoxetine, but the experimental duration was extended to 21 days using an environmentally relevant concentration of 20 ng/L in addition to the 20 μg/L from the earlier study. In contrast to the earlier work where already developed communities were exposed, the effect of fluoxetine on the process of biofilm colonization of stones was observed. It found that a trace concentration of 20 ng/L caused disruption of algal colonization and affected primary productivity on day 13 (Table 1). However, after 21 days, all parameters tested including chl-a, NEP and GPP were consistent with the control. These findings indicate that the colonization rate may be an important parameter when considering effects of pharmaceuticals on biofilm forming organisms. Robson et al. (2020) confirmed these observations, demonstrating that 20 ng/L fluoxetine causes a reduction in rates of gross primary production and community respiration in successional biofilm, while having no significant effect on established biofilm (Table 1).

Yang et al. (2019) evaluated the effect of another SSRI, sertraline, on lake microorganisms using microcosms. A 15-days exposure to 50 μg/L resulted in reduced chlorophyll-a content in the microcosms, suggesting this compound affected photosynthetic processes. Additionally, it was shown that sertraline can stimulate the growth of some organisms (bacteria) while reducing the diversity of others (cyanobacteria) (Table 1). Cui et al. (2021) conducted an extension of this study, by meta-transcriptomic profiling of functional variation of a microbial community affected by sertraline (50 μg/L). The analysis showed that sertraline disrupts the function of both eukaryotic and prokaryotic organisms, but the effects are more intense in bacteria (Table 1). Molecular studies have shown that sertraline causes inhibition of pathways related to lipid metabolism, energy metabolism, membrane transport function, and genetic information processing in the aquatic microbial community (Cui et al., 2021). Although the endpoints analysed in the work of Yang et al. (2019) and Cui et al. (2021) were different from those analysed for fluoxetine (Richmond et al., 2016; 2019; Robson et al., 2020), all of these studies confirm that psychoactive drugs can have significant effects on both the ecological structure and biochemistry of freshwater microbial communities.

Lawrence et al. (2005) analysed the effects of 10 μg/L ibuprofen on riverine biofilm, showing that this compound can significantly reduce cyanobacterial biomass (Figure 2) and affect bacterial community composition. The effects of ibuprofen, individually and in mixture with 17α-ethinylestradiol (103 and 148 mg/L, respectively) on biofilm were also analysed by McClean and Hunter (2020). Their study showed that this compound can decrease biofilm respiration and interfere with net primary production (Table 1). While respiration was also suppressed in the mixture, this was not the case for primary production. Although significant effects were observed in this study, the concentrations used were well above environmentally relevant values, which are usually measured in ng/L (Zuo et al., 2006; Rocha et al., 2013). Lawrence et al. (2007) examined the effect of another NSAID, diclofenac, on the microbial community. The observed effects at concentrations of 10 and 100 μg/L were inconclusive, as this compound had a stimulatory effect by increasing bacterial biomass but also had a negative effect on biodiversity. The observed changes were also different depending on season.

Kergoat et al. (2021) exposed freshwater biofilms for 4 weeks to two sulphonamides, sulfamethazine and sulfamethoxazole, at concentrations of 500 and 5,000 ng/L. Exposure to these antibiotics caused changes in structure, diversity, viability, and integrity of diatom cells. Additionally, a twofold increase in mortality of diatoms was observed in the sulfonamide-exposed group compared to the untreated biofilm. Sulfamethazine caused a reduction in species diversity as well as teratologies (deformities) in diatoms (Table 1). The observed changes indicate that sulfonamides pose a serious threat to the microbial community. Using mesocosms, Corno et al. (2014) examined the effects of a mixture of three commonly used antibiotics in Europe (tetracycline, imipenem, levofloxacin at 12.5 μg/L and 125 μg/L) on aquatic bacterial communities from European lakes. The results showed that the antibiotics caused a reduction in the number of bacteria (by about 75%), but this was not dependent on the drug concentration. In addition, the antibiotic mixture also had a large effect on the phenotypic distribution. In the presence of antibiotics (regardless of concentration), the bacteria formed large aggregates consisting of several different strains and small clusters composed of a single strain. It was proposed that the formation of aggregates is a resistance strategy developed by bacteria to create antibiotic-free space inside the clusters. Quinlan et al. (2011) observed that tetracycline (.1–100 μg/L) reduced microbiome numbers over a 28-days exposure period. Brosche and Backhaus (2010) exposed a community of limnic microorganisms to five antibiotics that inhibit protein synthesis: Streptomycin, chloramphenicol, fusidic acid, rifampicin, and chlortetracycline (in mixture and individually). The single-substance toxicity tests showed that each of the tested antibiotics inhibited bacterial protein biosynthesis, but with different potency: EC50 values ranged from 66 μg/L (chlortetracycline) to 46 mg/L (streptomycin) (Table 1). The adverse effects of levofloxacin and another of the tetracycline group antibiotics, oxytetracycline, on aquatic microorganisms were also demonstrated by Zhou et al. (2020). After 2 weeks of incubation, the microbial community was exposed to levofloxacin and oxytetracycline (individually) at a concentration of 5 μg/L. After 14 days exposure, the composition of the prokaryotic microbiota significantly changed at the genus level. These changes were dependent on the type of antibiotic, suggesting that the sensitivity of the bacterial community in this environment to a given contaminant varies. In the case of eukaryotes, 14-days exposure to levofloxacin and oxytetracycline (5 μg/L) did not significantly affect their diversity. The observed changes in the composition of the prokaryotic biofilm demonstrates that even low doses of these antibiotics can disrupt development, and generate phenotypic and genotypic variation, which in turn may accelerate the spread of resistance genes. Wang et al. (2022) showed in a 12-days mesocosm experiment that the lowest dose of oxytetracycline hydrochloride (5 mg/L) stimulated photosynthesis and biomass growth, while these parameters were inhibited at higher concentrations (25 and 75 mg/L). However, it should be noted that these concentrations are much higher than environmentally relevant. Effects on photosynthesis were also observed by Deng et al. (2022) when riverine biofilm was exposed to ofloxacin at concentrations of .01, .1, 1, 2, and 5 mg/L. After a 4-days exposure, concentrations of Chlorophyll-a and Chlorophyll-b decreased by 31.4% and 36.8%, respectively, compared to the control group, indicating ofloxacin at these “high” concentrations may result in reduced photosynthetic capacity.

Several studies have examined the effect of selected antibiotics on nutrient cycling. Gray and Bernhardt. (2022) treated microbial communities from stream sediments with three antibiotics (sulfamethoxazole, danofloxacin, and erythromycin as mixtures and individually at 10 μg/L) to measure rates of bacterial respiration and nitrogen transformation. Contrary to expectations, there was no effect of individual, and mixtures of, antibiotics on NO₃ ⁻ uptake or production of N₂O, N₂, CH₄, and CO₂. However, the mixture induced a marked increase in NH₄ ⁺ concentration. Rico et al. (2014) observed a potential effect on nitrification of the antibiotic enrofloxacin, added to microcosms for 7 days at nominal concentrations of 1, 10, 100, and 1,000 μg/L. The highest concentration of enrofloxacin significantly reduced the number of ammonia-oxidizing bacteria in the sediment, which translated into a higher ammonia concentration in the water and a lower nitrification rate (Table 1). Hou et al. (2015) showed that sulfamethazine causes a large decrease (approximately 20%–30%) in denitrification rate by inhibiting the growth of denitrifying bacteria (Table 1). This decrease was observed at drug concentrations up to 5 μg/L, above which denitrification rates remained constant. Inhibition of denitrification may lead to excessive accumulation of nitrogen (as ammonia) in the ecosystem and thus cause eutrophication. The denitrification process occurs through a series of related reduction steps mediated by various enzymes and genes. Sulfamethazine was found to increase N₂O production due to inhibition of its reduction to free nitrogen. Sulfonamide-mediated inhibition of denitrification was also demonstrated by Xu et al. (2020), who studied the effects of sulfamethoxazole on denitrification and N₂O release in river sediments. Collected sediments were incubated for about 1 month in the lab to remove sulfamethoxazole residues. Depending on the parameter under investigation, the antibiotic was introduced into the slurry at concentrations ranging from 1 to 100 μg/L. A clear decrease in the denitrification rate (by as much as 96%) with increasing sulfamethoxazole concentration was observed (Table 1). Sulfamethoxazole also decreased the rates of anaerobic ammonium oxidation (“anammox”); the rate of anammox decreased with increasing concentrations of sulfamethoxazole. Furthermore, 2-ethylhexyl-4-methoxycinnamate (UVB blocker) enhanced the observed inhibitory effects of sulfamethoxazole. Sulfamethoxazole also decreased the expression levels of important denitrification genes (nirS and nosZ) (Table 1).

Selected antibiotics have shown marked effects on enzymatic activity and biofilm metabolism. In these studies, enzyme activity was most commonly measured during or after biofilm exposure to pharmaceuticals. Paumelle et al. (2021) found that 14 days exposure of microcosms (containing colonized alder leaves immersed in water) to sulfamethoxazole and sulfamethazine at 5 μg/L, significantly reduced leaf-associated microbial community biomass (especially between days 3 and 7), accompanied by an increase in β-glucosidase activity, which has been attributed to the bacteria’s stress response to antibiotics (Table 1). In contrast, no effect of the drugs was observed on the four other enzyme activities measured: Cellobiohydrolase, phenol oxidase, alkaline phosphatase and leucine aminopeptidase. A significant increase in the activity of β-glucosidase with a 14 days exposure to sulfamethoxazole (500 and 5,000 ng/L) was confirmed by Pesce et al. (2021). On the other hand, sulfamethazine caused an approximate 20% decrease in β-glucosidase activity, which, according to the authors, may reflect the greater toxicity of sulfamethazine compared to sulfamethoxazole or the greater sensitivity of microorganisms to this antibiotic. These differences indicate that caution should be exercised when formulating conclusions about the properties of a given group of antibiotics on the basis of studies performed for a single compound. Effects of antibiotics on enzymatic activity were also noted by Wang et al. (2019) following a 192 h microcosm exposure to florfenicol and ofloxacin introduced separately at concentrations of 10, 100, and 1,000 μg/L. Both antibiotics caused an up to 6.7 times increase in the activity of antioxidant enzymes (superoxide dismutase and catalase) compared to baseline values for the untreated control groups (Table 1). The activity of these enzymes increased with increasing antibiotic concentration. Moreover, it was shown that in biofilms devoid of extracellular polymeric substances (EPS), antibiotic distribution coefficients (expressed by ratio of sorption amount to concentration of antibiotics at equilibrium solution) were 3.2 and 6.5 times higher (for florfenicol and ofloxacin, respectively) compared to the control group. This may indicate that EPS can act as a protective barrier for biofilm and increase its resistance to antibiotics, as noted above (Robson et al., 2020). Pu et al. (2021) studied the effects of erythromycin (6 μg/L) on freshwater biofilms. The biofilm used in this study was cultured in reactors, in aquarium water, allowing the lab to replicate a freshwater environment. The metabolic data indicated that despite the presence of the drug, cell-to-cell communication pathways, amino acid metabolism and peptidoglycan biosynthesis remained unchanged. However, changes in lipid and linoleic acid metabolism were observed (Table 1).

Several studies have used exposure-induced changes in transcription and gene expression to evaluate the effects of antibiotics on biofilm. Yergeau et al. (2010) used meta-transcriptomic analysis to investigate the response of river biofilms to four antibiotics: Erythromycin, gemfibrozil, sulfamethazine, and sulfamethoxazole. Biofilm reactors were inoculated with water from two rivers of different trophic status and exposed to different doses of drugs (.5 or 1 μg/L). It was observed that drug exposure can induce both positive and negative effects on gene transcription in the biofilm. Erythromycin had a positive effect on guanosine tetraphosphate production, the level of which is associated with many cellular functions. Both sulfonamides induced changes in DNA and RNA polymerase, which are responsible for replication and transcription. Moreover, sulfamethazine also affected changes in the expression of genes related to the envelope and outer membrane (Table 1). The effect of gemfibrozil on genes related to lipid metabolism was also observed. In their subsequent study, Yergeau et al. (2012) used the same group of drugs, testing whether, in addition to changes in gene expression, these drugs would also induce changes in the community composition. Based on the 16 S rRNA gene amplicons, they found that exposure to these antibiotics caused only minor changes in the composition of the bacterial community. However, it was observed that these drugs can induce numerous functional changes in the communities. Their presence had a large effect on the expression of genes responsible for nitrogen, phosphorus and carbon cycles (Table 1). Consequently, this may have led to a shift in the rate of phosphorus and carbon cycling, but this was not investigated in this work. These results again confirm that antibiotics at environmentally relevant concentrations can have a major impact on biofilm function and lead to disruption of important ecosystem processes.

One of the few papers directly examining the effects of pharmaceutical contamination on microbes in the marine environment is from Peele et al. (1981). This study investigated whether a pharmaceutical waste dump 64 km north of Arecibo (Puerto Rico Trench, Atlantic Ocean) affected microorganism composition in surface waters. Surface water was collected from the dump site and from several stations located nearby (close to the Puerto Rico shore) and transported to the laboratory. Microbial colonisation experiments showed that the pharmaceutical dumping site significantly affected taxonomic composition of microorganisms from surface waters. Vibrio and Aeromonas spp. were the dominant marine microorganisms culturable on agar from stations closest to the Puerto Rico shore and near the pharmaceutical dump site. However, the abundance of Vibrio and Aeromonas spp. decreased on the pharmaceutical dump sites, with an increase in the abundance and dominance of Gram-positive organisms (staphylococci, micrococci, and bacilli), which were proposed to have been introduced with the pharmaceutical waste (Table 2). In addition, an increase in specific bacterial activity rates, the ratio of ¹⁴C-labeled amino acid uptake to the total number of bacteria, was observed near the dump site. According to Peele et al. (1981) these changes in specific activity may be related to the shift in taxonomic composition of the bacterial community.